This project is a multi-level and interdisciplinary study of the source of extreme accuracy in a neuronal oscillation. Specifically, cellular and network properties of the pacemaker nucleus of the weakly electric fish, Apteronotus leptorhynchus, will be investigated experimentally and by computational model. Apteronotus has been chosen for investigation due to its extreme accuracy among biological network oscillators. The nucleus as a whole is known to fire synchronously and extremely regularly at frequencies between 700 and 1700 Hz. In recent years, general features of the pacemaker nucleus have been described. Future progress toward quantifying these features and investigating the source of high accuracy in this oscillator will be greatly enhanced, for one, by the recent experimental success in dissociating the cells of the oscillator, and second, by the integration of computational methods in tackling the research proposed here. This project will address the following questions: Are individual cells endogenously active? Preliminary evidence suggests they are. If so, does each cell maintain a highly regular frequency? How is the strength of coupling, and the number of cells in the network related to the degree of regularity? Both experiments and computational models will be used to address these questions.